Many biological functions are under circadian control, including release of hormones and neuropeptides, temperature, blood pressure and heart rate, bone remodeling, muscle regeneration, synaptic plasticity, sleep and activity cycles. We previously reported that older community dwelling men and women with weak circadian activity rhythms have a higher mortality risk and are more likely to develop mild cognitive impairment and dementia. However, little is known concerning the causes of age-related changes in circadian activity patterns in older adults and how these changes directly influence physiologic decline independent of other features of aging. In order to advance current knowledge regarding the role of objectively measured circadian activity rhythms in aging, we propose to (1) derive a novel set of activity rhythm parameters to complement our current traditional activity rhythm variables to (2) examine how these rhythms are associated with longitudinal changes in mobility, strength, and cognition, and (3) examine whether rhythms identify older adults at greater risk of incident or worsening disability, and furthermore whether these relationships are independent of concurrent changes in mobility, strength, and cognition. These analyses will be performed using existing data from two large (n>6,000), population-based, longitudinal cohorts with multiple-day actigraphy measured at two time points. We propose three study aims that will advance current knowledge regarding the role of circadian activity rhythms in aging. For our first specific aim we will derive a novel set of activity patterns using functional principle components analysis (fPCA) to complement the existing traditional (cosinor) activity rhythm variables in the Study of Osteoporotic Fractures (SOF) and Osteoporosis in Men (MrOS) cohorts. In our second aim we will examine how traditional and novel fPCA-derived activity rhythms are associated with longitudinal changes in walking speed, strength, and cognitive function. For our third aim we will test whether older men and women with disrupted circadian rest-activity rhythms are at greater risk of incident or worsening disability over time and whether these associations are independent of concurrent changes in walking speed, strength and cognitive function as well as comorbidities, and inflammatory and metabolic biomarkers. Activity rhythms are potentially modifiable behaviors and gaining a better understanding of the relationship between activity rhythms and subsequent age-related changes in mobility, strength, and cognitive function as well as incident disability may lead to novel interventions that improve health and well-being in the elderly. Furthermore, given the rapid growth of consumer-based devices to track activity, findings may prove useful in identifying older adults to target for intervention.